Download MARMORATAl - Journal of Neuroscience

Survey
yes no Was this document useful for you?
   Thank you for your participation!

* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project

Document related concepts

Caridoid escape reaction wikipedia , lookup

Mirror neuron wikipedia , lookup

Adult neurogenesis wikipedia , lookup

Molecular neuroscience wikipedia , lookup

Axon wikipedia , lookup

Electrophysiology wikipedia , lookup

Synaptogenesis wikipedia , lookup

Neural coding wikipedia , lookup

Stimulus (physiology) wikipedia , lookup

Subventricular zone wikipedia , lookup

Neuroregeneration wikipedia , lookup

Central pattern generator wikipedia , lookup

Axon guidance wikipedia , lookup

Multielectrode array wikipedia , lookup

Nervous system network models wikipedia , lookup

Clinical neurochemistry wikipedia , lookup

Synaptic gating wikipedia , lookup

Neuropsychopharmacology wikipedia , lookup

Pre-Bötzinger complex wikipedia , lookup

Premovement neuronal activity wikipedia , lookup

Development of the nervous system wikipedia , lookup

Circumventricular organs wikipedia , lookup

Neurotoxin wikipedia , lookup

Optogenetics wikipedia , lookup

Neuroanatomy wikipedia , lookup

Feature detection (nervous system) wikipedia , lookup

Channelrhodopsin wikipedia , lookup

Transcript
0270-6474/83/0309-1746$02.00/O
Copyright
0 Society
for Neuroscience
Printed
in U.S.A.
The Journal
of Neuroscience
Vol. 3, No. 9, pp. 1746-1759
September
1983
THE EXPRESSION
OF ANTIGENS
BY EMBRYONIC
NEURONS
AND GLIA IN SEGMENTAL
GANGLIA
OF THE LEECH HAEMOPIS
MARMORATAl
EDUARDO
* Department
R. MACAGN0,*,2
of Biological
RANDALL
R. STEWART,*
AND
BIRGIT
ZIPSERS
Sciences, Columbia University, New York, New York 10027 and $ Cold Spring
Cold Spring Harbor, New York 11724
Received
December
13, 1982; Revised
February
28, 1983; Accepted
March
Harbor
Laboratory,
28, 1983
Abstract
Monoclonal antibodies (mAbs) raised against adult leech nervous systems were screened on
embryos of the leech Haemopis
marmorata
in order to determine when in development specific
antigens are first expressed and the order in which they are expressed by different cells or tissues.
Three of the mAbs produced by Zipser and McKay (Zipser, B., and R. McKay (1981) Nature 289:
549-554) were screened: Lan3-1, Lan3-5, and Lan3-6. Each mAb shows a different pattern of
labeling in the adult leech nerve cord (Zipser, B. (1982) J. Neurosci. 2: 1453-1464). The embryonic
stages studied were from 5 days after egg deposition to 30 days (emergence from the cocoon). The
pattern of labeling was assayed in whole mounts using horseradish peroxidase-conjugated second
antibodies. The principal results are as follows. (1) Antigens recognized by Lan3-5 are first expressed
by the glia of the roots of the anterior segmental ganglia at 6 to 7 days, several days later by the
interganglionic connective glia, and near the end of embryonic development by ganglionic neurons.
An anterior to posterior temporal gradient is observed in the expression of these antigens. In
addition, Lan3-5 also labels the protonephridia and nephridia from early development onward. (2)
Antigens recognized by Lan3-6 are first expressed by a pair of neurons in each segmental ganglion
and later in development by additional neurons. By the time of emergence, however, only about
half of the neurons that label in the adult have done so, implying that some neurons express these
antigens postembryonically. Labeling with Lan3-6 is first seen in neuronal somata and only later
in neuronal processes. (3) Antigens recognized by Lan3-1 and expressed by segmentally specific
neurons in ganglia 5 and 6 are not detectable during embryonic development, but are so at early
postembryonic stages. Thus, these three mAbs provide an approach to study different aspects of
the development of the leech nervous system, specifically the relation between glial and neuronal
differentiation and the genesis of segmentally specific phenotypes.
The recent demonstration by Zipser and McKay (1981)
that, among monoclonal antibodies (mAbs) raised
against whole adult leech central nervous system (CNS),
some label identified cells or small groups of cells, provides direct evidence for the existence of individual molecular signatures in adult leech neurons and glial cells.
Little is currently known about the roles of the antigens
bound by these mAbs in establishing the structure, function, or synaptic connectivity of these cells. Understandi We thank Carol Schley for her invaluable
technical
assistance
and
Murray
Flaster
for his careful reading
of the manuscript.
This work
was supported
by National
Institutes
of Health
Grants
NS 14946 (E.
R. M.) and NS 17984 (B. Z.), by National
Science Foundation
Grant
BNS 78-248-72
(B. Z.), and by a Whitehall
Foundation
grant.
* To whom correspondence
should be addressed,
at Department
of
Biological
Sciences, 1003 B Fairchild,
Columbia
University,
New York,
NY 10027.
1746
ing these roles requires information about the identity,
localization, and temporal expression of these antigens.
Initial efforts to identify them and to determine their
distributions within cells have been reported recently
(Zipser et al., 1981, 1982; McKay et al., 1982; Hockfield
and McKay, 1983). We have begun to examine when
during the life of the animal these antigens first appear,
and what the degree of cell differentiation is when the
antigens are first expressed.
A practical reason also stimulates our interest in studying the developmental pattern of labeling with the mAbs
raised to adult tissues. Whereas particular neurons can
be relatively easy to identify in an adult nervous system
on the basis of specific morphological and electrophysiological criteria, such identification may be difficult, if
not impossible, during neuronal ontogenesis. The reasons for this are many. For example, size is often used
to distinguish a cell from its neighbors, but at early
The Journal
of
Neuroscience
Labeling
of Embryonic
Leech Neurons
developmental stages neurons are frequently similar in
size. Furthermore, the characteristic shape and electrical
properties may be very different in embryonic and adult
stages. Markers that distinguish individual neurons from
each other as well as from non-neuronal cells would thus
be an invaluable aid in the study of the early stages of
the differentiation of these neurons. To the extent that
neurons express individual molecular signatures early in
their development, antibodies raised against such molecules can play a useful role as identifying markers.
The adult leech CNS consists of a head ganglion (with
supra- and subesophageal sections), 21 segmental ganglia
(SG), and a tail ganglion. Paired lateral connective
nerves, along with an unpaired medial one (Faivre’s
nerve), run along the nerve cord from the head to the
tail ganglia. The SG are connected to the periphery by
pairs of bilateral nerve roots. During development 32
ganglionic primordia are made. The four most anterior
fuse to form the subesophageal section of the head ganglion, the seven most posterior fuse to form the tail
ganglion. The 21 SG are quite similar to one another,
each containing about 400 neurons, with the exception
of those in the 5th and 6th body segments, which contain
a few hundred more cells in hirudinid leeches (Macagno,
1980).
The work reported here was carried out on the hirudinid leech Haemopis marmorata. The general development of hirudinid leeches has been described by many
authors (reviewed in Dawydoff, 1959 and Mann, 1962).
A recent report by Fernandez and Stent (1983) describes
in some detail the formation of the germinal plate and
CNS of Hirudo medicinalis. Our observations on Haemopis marmorata indicate that these aspects of development are essentially identical in these two species.
We have examined the pattern of labeling of the SG
of H. marmorata at different developmental stages by
mAbs raised against adult nerve cords. The mAbs used
(Lan3-5, Lan3-6, and Lan3-1) are among those first
described by Zipser and McKay (1981). We selected these
three mAbs because in the adult they recognize antigens
with very different distributions (Zipser, 1982). Lan3-5
labels glial cells in the connectives and roots and between
10 and 20 neurons in each SG. With Lan3-5 we can
study the development of central and peripheral fiber
tracts and the relative time of expression of antigens by
neurons and glial cells. The second mAb, Lan3-6, labels
about 10% of the neurons in each SG in the adult, and
thus we can ask whether all neurons that express the
antigens recognized by this mAb do so at the same time
in development and, ultimately, whether they do so at a
specific point during their differentiation. Lan3-1 labels
a pair of small neurons in each SG and an additional
pair of larger neurons in ganglion 5 and ganglion 6, which
innervate the sexual organs. Using this mAb we can ask
when in development this segmental specialization occurs. The results presented in this paper answer these
and other questions. Some of these results have been
reported previously (Stewart et al., 1982).
Materials and Methods
Large, gravid Haemopis marmorata were placed in
plastic boxes containing a mixture of potting soil and
milled sphagnum moss. The boxes were checked every
by Monoclonal
Antibodies
1747
other day for cocoons containing embryos. Once found,
the cocoons were placed either in loo-mm Petri dishes
or in loo-ml plastic cups and covered with moist soil.
These were maintained in an incubator at 23 f 1°C.
Embryos of various ages were selected for dissection
and fixation. Generally, a cocoon was immersed in cold
saline (115 mM NaCl, 1.8 mM CaCl, 1.8 mM MgCl, 4.0
mM
KCl, and 10 mM Tris maleate adjusted to pH 7.4
with NaOH), one end was sliced off with a razor blade,
and the embryos were removed. While in saline, each
embryo’s yolk was removed and the embryo was pinned
out dorsal side up in Sylgard-covered dishes. Adult nerve
cords to be used as controls were also dissected and
pinned out in Sylgard-covered dishes. Both embryos and
adult nerve cords were fixed at room temperature in
Bouin’s fixative for 4 hr or in 4% paraformaldehyde in
0.1 M phosphate buffer for 30 min. After fixation they
were removed from the dishes and were either acetone
extracted for 7 min or dehydrated and xylene extracted
for 1 min, then stored in phosphate-buffered saline
(PBS) containing 1 drop of 10% sodium azide/5 ml or
incubated in mAb as described below.
Fixed and extracted tissues were placed in mAb supernatant or ascites fluid (dilution of 9 parts supernatant
to 1 part 20% Triton X and of 1 part ascites to 1000
parts PBS/2% Triton X). After overnight incubation
with mAb, the tissue was rinsed for 5 min with PBS/B%
Triton X, incubated 2 to 4 hr in horseradish peroxidaseconjugated goat anti-mouse IgG (heavy and light chain
specific; N. L. Cappel Laboratories), and rinsed again
with PBS/B% Triton X for 5 min. Next, the tissue was
repinned in Sylgard dishes, rinsed briefly in 0.05 M Tris
buffer, and incubated in 20 mg of diaminobenzidinel5
ml of 0.05 M Tris buffer for 5 to 15 min, then, 1 drop of
1% H,Oz was added, and the reaction was allowed to
proceed until the tissue became light brown. Lastly, the
tissue was dehydrated in an ethanol series, cleared in
xylene, and mounted using Permount between two coverslips.
The degree and the reproducibility of labeling with
each mAb were dependent on the type of fixative used.
For example, Lan3-1 only labeled tissues fixed in
Bouin’s, whereas Lan3-6 labeled optimally when paraformaldehyde was used. Lan3-5, however, worked equally
well with either method of tissue fixation. The embryonic
and adult labeling patterns were always obtained for
each mAb using the same procedures.
mAb staining patterns were analyzed in whole mounts
under Nomarski optics using a computer-coupled light
microscope (described in Macagno, 1980). The outline of
a ganglion and the perimeter of each stained neuron were
drawn into the computer to determine the number and
spatial distribution of neurons labeled by each mAb. For
Lan3-6, the labeled cells in each SG were counted twice,
once each by different investigators, and then averaged
to obtain the values quoted in this report. Nomarski
optics were also used to photograph the embryos and to
study their morphological features to assesstheir relative
ages.
Results
The observations reported below were made for the
most part on embryos from 5 days old to emergence,
1748
Macagno et al.
ii
I
Vol. 3, No. 9, Sept. 1983
which occurs about 30 days after egg deposition at temperatures of 22 to 24°C. The general level of development
of the nervous system at different embryonic stages is
described in Figure 1. Comparison of the developmental
staging scheme proposed by Stent et al. (1982) for the
embryonic development of the glossiphoniid leech Helobdella triserialis
with our own observations on the hirudinid Haemopis
marmorata,
suggests to us that the embryonic stages we have studied correspond to those that
these authors labeled stages 8 through 11. Although.the
early developmental stages in these two leech orders are
very different (Dawydoff, 1959; Mann, 1962), it is probably the case that later stages, and particularly the
formation of the CNS, are fairly similar (Fernandez and
Stent, 1983). In the studies with Lan3-1, some early
postembryonic stages were also examined.
The pattern
of label obtained with Lan 3-5. In the adult
CNS, Lan3-5 has been shown to bind to glial components
of the roots and connectives and to a small number of
neurons (10 to 20) in each SG, of which two pairs are
the pressure mechanosensors (P cells).
The first structures in the embryonic CNS that are
labeled by Lan3-5 are the ganglionic roots. Little labeling
is detected in 6 to ‘I-day embryos (Fig. 2, a and b), even
in anterior segments. At later stages (Fig. 2, c and d),
more posterior roots become labeled and those that are
anterior are labeled more intensely. In addition, label
extends further along the roots toward the body wall,
and a cell body lying between the roots at the ganglionic
margin (probably of the glial cell of the roots) becomes
intensely labeled (see Fig. 3, a and b). The pattern of
labeling indicates that Lan3-5 is binding to a cytoplasmic
component of the glia that ensheath the neuronal processesin the root nerves. A temporal gradient of maturation from the older anterior segments to the younger
Figure 1. Drawings of Haemopis embryos of various ages.
The numbers under each drawing refer to the number of days
after cocoon deposition. Only the area of the germinal plate
has been drawn; the larval organs are not shown, with the
exception of the first four stages,where the larval mouth is
indicated by an arrowhead.
All the ganglionic primordia are
visible by 8 days. The four gangliathat form the subesophageal
ganglion begin to fuse at 8 days, the seven that form the tail
ganglion begin to fuse at 10 days. The tail sucker begins to
form at 10 to 11 days. Anterior is up.
Figure 2. Patterns of labeling with Lan3-5 in whole mounts of various embryonic stages.In all casesanterior is up. a, Low
power micrograph of a 6- to ‘i-day embryo. The picture extends from the first SG (I, partially cut off at the top) to the ninth SG
(9, partially cut off at the bottom). The forming nephridia are indicated on the left side by arrowheads. Labeling with Lan3-5
can be found in the four most anterior nephridia. Bar = 100 pm. b, Higher magnification micrograph of a region of a. The region
includes SGl to SG4 (I to 4). Labeling with Lan3-5 can be seenin the nephridia (N) and in the lateral margins of the first
ganglion, where the root glia are beginning to form (r). Longitudinal processesare indicated by arrowheads in ganglia 3 and 4.
The ganglia lie adjacent to one another, and the connectives have not formed. Bar = 50 pm. c, Micrograph of a central region
(SG8 to SG12) in a 12-day embryo. The nephridia (N) are heavily labeled by Lan3-5, as are the ganglionic roots (r). The
interganglionic connectives (c) are visible, and the ventral blood sinus (bs) is beginning to form. Magnification is the sameas in
a. d, Micrograph of the posterior region of a 13-day embryo, including SG13 through SG20 (only 13 and 20 are numbered). The
labelednephridia are seenon each side of the nerve cord. The ganglionic roots are labeled in all ganglia,but label extends further
toward the periphery in the more anterior ganglia. Bar = 100 pm.
The Jou .rnal of Neuroscience
Labeling of Embryonic Leech Neurons by Monoclonal Antibodies
Figure 2
1749
Macagno
et al.
Vol. 3, No. 9, Sept. 1983
Figure 3. Labeling with Lan3-5 in a whole mount of a 25day embryo. Anterior
is up in all micrographs.
a, Low power
micrograph
showing SG15 to 17. The roots are labeled, as are the nephridia
on the right. The blood sinus (bs) is well formed at
this stage and contains blood. Bar = 100 pm. b, Detail of the roots of SG9. The labeled cell between the roots contains a nucleus
(n). Bar = 20 pm. c, SGl5 and a portion of the connective anterior to the ganglion. The connective and the roots are heavily
labeled, but neurons are not. bs indicates the blood sinus. Bar = 20 pm. d, Region of the labeled connective posterior to SG12.
The glial nuclei are indicated (n). Magnification
is the same as in c.
The Journal
of
Neuroscience
Labeling of Embryonic Leech Neurons by Monoclonal Antibodies
posterior ones is apparent. It appears, therefore, that
Lan3-5 binds to an antigen which is expressed early in
the differentiation
of the root glia, at a time when the
initial connections between the SG and the periphery
are being made.
In contrast to the roots, binding of Lan3-5 to the
connectives is not detectable until about 25 days of
development (Fig. 3, c and d), more than 2 weeks after
neuronal processes are first seen in the connectives. The
distribution
of label, particularly
where the connectives
enter a ganglion, indicates that Lan3-5 is labeling the
two giant glia which ensheath the neuronal processes
between each pair of adjacent SG. Since these glia can
be seen considerably earlier in development than 25 days
(we first see them in whole mounts at 11 to 12 days), it
would seem that expression of the antigens bound by
Lan3-5 occurs at relatively different stages in the differentiation of the connective and root glia. In comparison,
the six packet glia and the two neuropil glia are clearly
seen in each SG by the end of embryogenesis, but show
no evidence of labeling with Lan3-5, in agreement with
the lack of such labeling in the adult.
Labeling of neuronal somata by Lan3-5 is not detected
until late in embryogenesis, between 25 and 30 days, just
prior to hatching. Even then, labeling can only be seen
clearly in some cell bodies in the supraesophageal ganglion, and only faintly, if at all, in a few cells in the SG.
By this time P cells can be identified by size and location,
but only in a few SG do we detect any binding of Lan35 to these neurons. Thus, the neuronal Lan3-5 antigens,
be they identical to or different from the glial antigens,
are for the most part expressed at a detectable concentration postembryonically.
TABLE
1751
In addition to labeling nervous tissue, Lan3-5 labels
the embryonic protonephridia
and nephridia. Labeling
of the nephridia was first observed in 6- to 7-day embryos, soon after the nephridial primordia are apparent
under Nomarski optics (see Fig. 2, a and b). Labeling is
stronger in the more anterior segments, disappearing
gradually more posteriorly, although nephridial primordia are visible (Fig. 2~). At later stages (8 to 9 days), the
label extends further posteriorly and the anterior nephridia are more convoluted. By 11 to 12 days, all nephridia are labeled (Fig. 2, c and d). The binding of LanS5 to the nephridia of different body segments is a useful
marker of embryonic stages, particularly
between 6 and
12 days of development.
We do not know at present
whether neural and nephridial antigens are identical or
not.
The pattern
of label obtained with Lan3-6. In the adult,
Zipser (1982) reported finding that approximately
10%
of the nearly 400 neurons in each SG label with Lan3-6.
Some cells label heavily, others lightly. In order to present a quantitative
comparison of the adult and embryonic staining patterns, we counted the neurons stained
with Lan3-6 in the SG of two adult leeches. The data are
presented in the last column of Table I. An example is
shown in Figure 4. Some variability is found from preparation to preparation. To what extent this is a consequence of real differences in the expression of the antigen
or antigens bound by Lan3-6 is not known.
Examples of the labeling of ganglia by Lan3-6 at
different embryonic stages appear in Figures 5 to 7.
Counts of numbers of cell bodies labeled as a function of
embryonic age are presented in Table I. We first find
labeling by Lan3-6 at 8 days of development in a pair of
I
Numbers
of neurons labeled by Lan.F-6 in each segmental ganglion (SC) at various times during embryogenesis
and in adults
The times are given as days after cocoon deposition
(e.g., 8 d is 8 days). For each time, a number
(n) of embryos
was examined.
The
each SG are the average of separate counts by two investigators
for all specimens
at each time, rounded
off to the nearest one-half.
Days
Segmental
Ganglion
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
data for
of Development
Adult
8d
(3)
9d
(3)
2
2
2
1.5
2
2
1
0
1
0
0
0
0
0
0
0
0
0
0
0
0
2
2
2
2
2
2
2
2
2
1
1
1
0.5
0
0
0
0
0
0
0
0
10 d
(1)
11 d
(2)
12 d
(3)
I
2
2.5
2
2
2.5
2
2
2.5
2.5
1.5
2.5
2.5
3.5
1.5
2.5
2
1
0.5
0
0
8
8
9.5
8
7.5
8
5.5
7
7
5.5
4.5
4
3.5
4
4
2
2.5
2
1
1
16
18
15.5
13
12
15
15.5
15
12
10
10
9
10
8
9
I
6
4
2
1.5
13 d
(1)
16.5
19.5
16
18
19
20.5
26
21
21.5
18.5
17
17.5
17.5
15.5
17
16.5
14.5
17.5
13.5
10
9
25 d
(3)
19
20
16
22.5
18
25
22
21
17
23
28
17
16.5
15
15
20
15
13.5
(2)
34.5
35.5
38
36
39
38
50.5
54
50
47
51
47
41
46
48
44.5
43
47
41
35
31
1752
Macagno et al.
Vol. 3, No. 9, Sept. 1983
Figure 4. Labeling of an adult SG8 with Lan3-6. a, Micrograph of a whole mount. Many processes are labeled in the connectives,
roots, and neuropil. Anterior is up. Bar = 100 pm. b, Computer-aided drawing of the ganglionic boundary and the labeled neurons.
Heavily labeled cells are drawn with solid lines, more lightly labeled ones are drawn with dotted lines.
ventrally
located cells (see Fig. 5). A pair of cells in
approximately
the same location is labeled in the first 6
to 10 ganglia in &day animals, and as development
proceeds, a similar pair labels in more posterior ganglia,
until 11 days, when essentially all ganglia show this pair.
Additional cell bodies begin to label with Lan3-6 at 11
days. These appear initially in the anterior ganglia and
gradually, later, in those more posterior
(see Table I).
The next cells to label are a pair of dorsal cells, followed
by other dorsal and ventral cells (see Figs. 6 and 7).
Although neurons have not been identified directly in
any embryos, it is possible to tentatively identify by size
and position the Retzius and the P cells, which often
label with Lan3-6 in the adult. We see only very light, if
any, labeling of these neurons even at the latest embryonic stages we have examined.
In order to visualize more easily anteroposterior
differences in labeling with Lan3-6, the average numbers of
labeled cell bodies per ganglion for four groups of segmental ganglia were calculated. These averages, along
with the averages for all ganglia, are given in Table II
for each developmental
stage and for adults. They are
also presented graphically in Figure 8. The average number of labeled neurons per ganglion rises by 13 days of
development
to values that remain approximately
constant until the end of embryogenesis.
These values are
only about one-half of the adult values and, therefore,
additional neurons must label postembryonically.
Differences along the anteroposterior
axis are also apparent in
Figure 8. For example, the average numbers of labeled
neurons in the most posterior
SG remain below the
overall average throughout
development, a characteristic
which also prevails in the adult. The SG in midbody
regions (SG6-10, SGll-15)
begin to label with Lan3-6
after the most anterior ones (SGl-5) do. However, by 25
days they show more labeled neurons than do the anterior ganglia, as they do in the adult (see Table II).
The labeling of cell bodies with Lan3-6 is followed
within a few days by labeling of processes in the ganglionic neuropils (see Fig. 6). By focusing through a ganglion, the first processes to label in the neuropil can be
followed to the cell bodies of the neurons that labeled
earliest in that ganglion. As development proceeds, la-
beled processes are also found in connectives
and roots
(see Fig. 7), and more processes are labeled in the neuropils.
The pattern of label obtained with Lan3-1. In the adult,
Lan3-1 labels a pair of posterior dorsal cells in each SG,
an additional pair of lateral ventral cells in ganglia 5 and
6, and processes and varicosities in the ganglionic neuropils whose cells of origin are unidentified (Zipser,
1982).
The youngest embryos in which binding of Lan3-1 was
detected were 11 to 12 days old. A pair of dorsal cell
bodies, located near the posterolateral margins of each
SG, were labeled, along with a few short processes and
varicosities in the central region of the neuropil. No
binding of Lan3-1 to processes of the labeled neurons
was detected. The bound Lan3-1 was distributed within
the cell bodies in a non-uniform manner, being more
concentrated in the region surrounding the cell nucleus.
No additional cells were found to be labeled in either
SG5 or SG6. At later embryonic stages more processes
and varicosities in the neuropil, some nerve fibers in the
connectives, and the initial segments of the processes of
the labeled pairs of neurons were found to bind Lan3-1.
However, in no embryonic stage up to 25 days did we
find additional cells in SG5 or SG6 with detectable
binding of Lan3-1. The locations of the labeled embryonic cells are similar to those of the cells common to all
SG described by Zipser and McKay (1981) for the adult
nerve cord. The labeled cells in SG6 from a 20-day
embryo can be seen in Figure 9.
In order to determine whether the lack of Lan3-1
labeling of the extra neurons in SG5 and SG6 is due
either to absence of the cells before hatching or to their
failure to express the antigens bound by Lan3-1, we
counted the cells in individual SG at various embryonic
and a few early postembryonic stages. Preliminary data
on SG5, SG6, SG7, and SGlO from a few specimens
indicate that cell numbers in each of these ganglia are
approximately the same in the embryo. At 12 days there
are about 440 ceils per SG, at 20 days only about 395.
Since the adult number of neurons in SG7 and SGlO is
about 400 (Macagno, 1980), it would appear that the
adult value is achieved at late embryonic stages, although
The Journal
of Neuroscience
Labeling
of Embryonic
Leech Neurons
by Monoclonal
Antibodies
1753
Figure 5. Micrographs
of SG4 in a lo-day embryo labeled with Lan3-6. a, b, and c show three focal planes, from dorsal to
ventral. d is a computer-aided
drawing of the ganglionic boundary and the two neurons that show label at this stage. The labeled
neurons are indicated by arrowheads
(b and c). (These and subsequent micrographs
were taken with Nomarski optics in order to
highlight
details of the SG, but at the expense of some ambiguity in the visualization
of lightly labeled neurons. Bright-field
illumination
was employed in the determination
of which neurons were labeled by the mAbs.) Anterior is up. Bar = 25 pm.
we cannot exclude at this time the possibility that accurately matched cell proliferation
and cell death maintain
this number the same at later times. With respect to SG5
and SG6, however, it is clear that further cell prolifera-
tion must take place, since the adult has close to 700
neurons in each of these ganglia (Macagno, 1980). Preliminary cell counts in a 46-day animal (about 2 weeks
post-emergence) show an increase in cell number in SG5
1754
Macagno et al.
Vol. 3, No. 9, Sept. 1983
Figure 6. Micrographs of SG3 in a 13-day embryo labeled with Lan3-6. a, b, and c show three focal planes, from dorsal to
ventral. Labeled neurons in focus in these micrographsare indicated by arrowheads (someare not in focus in any of the three
micrographs).Labeledprocessesare also indicated (p). d is a computer-aideddrawing of the ganglionic boundary and the labeled
neurons.Heavily labeledcells are drawn in solid lines, lightly labeled onesare drawn with dashed lines. Anterior is up. Bar = 25
pm.
The Journal
of Neuroscience
Labeling of Embryonic Leech Neuronsby Monoclonal Antibodies
1755
Figure 7. Labeling with Lan3-6 in a 25-day embryo. a and b showtwo focal planesof SGl2. Many neurons, aswell asprocesses
in the roots, connectives, and neuropil, are labeled. c, Computer-aideddrawing of the ganglionic boundary and labeled neurons
in SGl2; heavily labeledcell bodiesare drawn in solid lines, lightly labeledonesare drawn in dotted lines. In a, b, and c, anterior
is up. Bar in a = 100 pm, and appliesalsoto b and c. A lower power micrograph of the body wall to the right of SG8 is shown in
d. Processesin the roots and their branchesare seento be heavily labeledby Lan3-6. The arrow points toward anterior; the bar
in d = 100pm.
(430 cells) and no significant change in SGlO (386 cells),
indicating that the additional neurons in the sex ganglia
begin to appear early postembryonically. In correlation
with this possibility, when we screened Lan3-1 on 46-
day-old animals, we detected a low level of binding on
the cell bodies of an additional pair of neurons in SG5
and in SG6. These were located on the ventral surface
(Fig. lo), in positions similar to those found in the adult
1756
Macagno
et al.
TABLE
Average
numbers
of neurons
per ganglion
Vol. 3, No. 9, Sept. 1983
II
labeled by La&-6
for grouped
segmental
ganglia at different
segmental ganglia in different
times in development
regions
of the nerve
cord and for all
The errors quoted are standard deviations of the average. The primary data appear in Table I.
Days of Development
Segmental
Ganglion
8d
9d
Gl to G5
G6 to GlO
1.9 + 0.2
0.8 + 0.8
Gil to G15
G16 to G21
0
2.0 f 0
1.8 k 0.4
0.5 + 0.5
0
2.1
2.3
2.3
1.0
1.0 + 0.9
1.9 f 0.9
0
Gl to G21
0.65 f 0.9
10 d
f
+
+
*
11 d
0.2
0.3
0.8
1.0
8.1
7.0
4.3
2.1
+
f
-c
+
Adult
12 d
0.9
0.9
0.8
1.1
15.6
13.9
9.4
4.9
5.2 + 2.6
f
k
+
f
13 d
2.0
1.7
0.9
2.9
10.4 + 4.7
17.8
21.5
16.5
13.5
+
-c
f
+
25 d
1.5
2.8
0.8
3.4
17.2 f 3.7
18.3
21.7
20.3
15.7
f
f
+
+
2.1
2.5
5.1
2.5
37
48
47
40
19.1 + 3.9
+
+
+
-+
2
6
4
6
43 -c 6
50
.._....
0
. .
.
----a--
40
SG,-5
SG6-10
----A-.-SGll-15
-..--.
_.__ .SG16-21
-SGl-21
30
NUMBER
OF
LABELED
CELLS
20
10
o,..__~?~~-~..!~...”
o--
A
A
6
7
8
9
10
DAYS
11
OF
12
13
h
25
ADULT
DEVELOPMENT
Figure 8. Graph of the average number of neurons labeled by Lan3-6 in various groups of segmental ganglia as a function of
days in development.
The numerical data are presented in Table II. Only the standard deviations of the averages of all ganglia
(SGI-21) have been indicated in the graph. Symbols are displaced laterally for clarity.
by Zipser and McKay (1981). We do not know at present temporal features which we will discuss for each mAb
whether they are the same cells, though it would seem individually.
likely.
Lan3-5.
The antigens recognized by this mAb are
expressed in the embryo by both nervous tissue and
Discussion
nephridia. The uniform distribution of label throughout
The data presented show that all three mAbs raised the nephridia suggests that Lan3-5 does not recognize
against adult leech nerve cords label embryonic struc- neural components in this tissue. Whether the molecules
tures. The patterns of labeling have distinct spatial and that are recognized are the same or only share similar
The Journal
of Neuroscience
Labeling of Embryonic Leech Neuronsby Monoclonal Antibodies
i
__ -,
,:./,r ,* .*’
1757
Figure 9. Micrographs of SG5 from a ZO-dayembryo labeled with Lan3-1. a, b, and c showthree focal planes from ventral to
dorsal. Only two small neurons are labeledby Lan3-1 at this stage (arrowheads).
The male genital pore is indicated by an arrow
in a. d showsa drawing of the ganglionic boundary and the two labeled cell bodies.Anteripr is up. Bar = 50 pm.
1758
Macagno et al.
Vol. 3, No. 9, Sept. 1983
Figure 10. Labeling with Lan3-1 in a 46-day animal (2 weeks post-emergence). The ganglion shown is SG6. Four neurons are
labeled, two located dorsally and two ventrally. This is the earliest stage that we find the ventral cells labeling. a, Micrograph of
The label is most visible in the
a whole mount of SG6. One of the ventrally located neurons is seen at the right (arrowhead).
perinuclear region. The other labeled ventral neuron is not seen at this plane of focus. The pair of labeled dorsal neurons common
to all ganglia appear out of focus (arrowheads). b, Computer-aided drawing of this ganglion. The four neurons labeled by Lan31 are outlined. The locations of the centers of each of the nuclei of the 430 neurons in this ganglion are indicated by points.
Anterior is up.
antigenic sites is unknown, but efforts to characterize
the Lan3-5 antigens are presently under way (M. Flaster
and B. Zipser, personal communication).
It is of interest
to note that two other mAbs raised against leech CNS
(Lan3-2 and Laz2-369), which bind to both nervous and
gut tissues, react with multiple antigen bands on Western
blots prepared from these tissues (Zipser et al., 1982;
Hogg et al., 1983). In these instances the CNS neuronal
antigens and the gut epithelial antigens are different. In
any case, the labeling of the nephridia by Lan3-5 provides
us with a useful marker for staging embryonic ganglia in
a manner that is independent of the nervous system
itself.
Within nervous tissue, the first structures labeled by
Lan3-5 are associated with the forming roots. The distribution of label indicates that the cells that bind Lan35 are glia and that neuronal processes in the roots are
not themselves labeled. It is of interest to note that the
labeling with Lan3-5 allows us to identify these presumptive root glia at stages (6 to 7 days) when processes
of central neurons are beginning to grow out to the
periphery (Wallace, in Weisblat, 1981) and when peripheral neurons are beginning to grow into the CNS (unpublished observations). These glia may be playing a role
in the establishment
of peripheral nerves, a possibility
that can be studied by labeling with Lan3-5 and dyefilling individual
neurons in the same specimen. This
would allow us to visualize simultaneously
the axonal
and glial processes to determine whether they grow at
the same time or whether one precedes the other.
The labeling of the interganglionic
connective glia
occurs late in embryogenesis,
at a time when many
processes are readily recognized in the connectives. In
this case, in contrast to the case of the roots, the antigens
are expressed considerably later than the time when the
glia first appear. Labeling of neurons by Lan3-5 also
occurs late in embryogenesis,
at a time when many
neurons have extensive dendritic fields. The P cells, for
example, are among the earliest to differentiate (J. Kuwada and A. P. Kramer, submitted
for publication);
however, they label with Lan3-5 only at the end of
embryogenesis or postembryonically.
No simple temporal correlation
between the expression of antigens
recognized by Lan3-5 and specific types of morphologically recognizable events is apparent from these initial
studies.
La&-6. The expression by ganglionic neurons of the
antigenic sites recognized by Lan3-6 is a gradual process.
For a particular SG, as shown by the data in Table I, the
number of neurons labeled by Lan3-6 increases monotonically with embryonic age, with about half of the
number in the adult labeling at the end of embryogenesis.
The significance of this should be considered in light of
our preliminary
observations (unpublished)
concerning
cell number as a function of embryonic age. We have
found that SG in lo- to 12-day embryos have a significantly larger number of neurons than adult SG (except
in SG5 and SG6. The number decreases to the adult
value by the end of embryogenesis. This net loss could
be due entirely to cell &ath or to a combination
of cell
The Journal
proliferation
number
of Neuroscience
Labeling
of Embryonic
Leech Ne !urons by Monoclonal
and cell death. Whatever is the case, the
of cells labeled by Lan3-6 is increasing
at the
same time that a significant
loss of cells (up to 15%) is
taking place in each ganglion.
Although
no loss of cells
labeled by Lan3-6 is clearly evident, the loss of a few
such cells would probably
be hidden within the variability
we see from specimen to specimen. It is of interest to
note that Stuart et al. (D. K. Stuart, S. S. Blair, and D.
A. Weisblat, submitted for publication) have found more
neurons that contain monoamines
at some embryonic
stages than are found in the adult. Whether
this is due
to a loss of some of these neurons or to their stopping to
make
monoamines
Segmental
labeling
is not known
differences
with
Lan3-6
and
at present.
are seen in both the onset of
the total
number
of cells
beled. The time dependence seems to be a reflection
general
temporal
gradient
of development
from
la-
of a
anterior
to posterior which is seen in both neuronal and other
tissues. Segmental differences in the numbers of labeled
neurons seen in the adult (see Table II) begin to appear
in the middle of embryogenesis, at around 13 days of
development. This seems particularly
clear in the more
anterior ganglia, which in the adult show fewer neurons
labeled with Lan3-6 than do the ganglia in the middle of
the nerve cord. This observation supports the suggestion
that segmental differentiation
may be an early event.
However, as will be discussed below with respect to Lan31, we have reason to believe that this hypothesis does
not apply to all segmental differences.
Among the earliest neurons labeled by Lan3-6, the
label is first found on the cell bodies, and only later in
processes. We can not determine from our observations
whether this is due to these neurons lacking processes at
these times or not. However, it is the case that certain
neurons that label at later times, particularly
postembryonically, do so after they have undergone considerable
morphological
differentiation
(e.g., the P cells). An attractive, though entirely hypothetical possibility, is that
expression of the antigens by different neurons is an
inductive phenomenon, occurring serially as the various
neurons involved each become part of a particular functional circuit.
Lad&l. At the embryonic stages in which the labeling
of neurons with this antibody was studied, only a homologous pair of dorsal cells was found to be labeled in
all ganglia. By 2 weeks after emergence, however, additional cells in the sex ganglia begin to express detectable
levels of antigens bound by Lan3-1. The ventral pairs of
cells labeled by Lan3-1 in ganglia 5 and 6 of adults
(Zipser, 1982) are part of the additional complement of
a few hundred cells that are found only in these two
ganglia (Macagno, 1980). Preliminary
data from cell
counts at late embryonic and early postembryonic stages
indicate that these additional
cells do not appear in
ganglia 5 and 6 until postembryonic
stages. Since in one
case (ganglion 6) the extra pair of cells labeled by Lan31 are known to have a function in sexual behavior (Zipser
1979a, b; Zipser and McKay, 1981), the delay in the
appearance of these neurons may reflect the delay in
maturation of sexually related structures in these leeches.
Antibodies
1759
Whether these neurons arise from delayed mitosis of
intraganglionic
precursor cells or from the migration of
cells into these ganglia (as seen in Aplysia by M. Jacob,
submitted for publication)
remains to be determined. In
contrast
to the segmental
differences
found
in the embryo, the segmental specialization
Lan3-1
occurs
postembryonically.
The
with
Lan3-6
detected with
extra
pairs
of
ventral neurons are labeled by Lan3-1 soon after they
first appear in the sex ganglia. Thus Lan3-1 will be a
useful marker for identifying and studying these specific
neurons early in their differentiation.
References
Dawydoff, C. (1959) Ontogenese des annelides. C. Developpement des Hirudinees.
In Traite de Zoologie. Tome V: Annelides, P. Gras&, ed., pp. 665-682, Masson et Cie, Paris.
Fernandez, J., and G. S. Stent (1983) Embryonic development
of the hirudinid
leech Hirudo medicinalis:
Structure, development and segmentation
of the germinal plate. J. Embryol.
Exp. Morphol., in press.
Hockfield,
S., and R. McKay (1983) Monoclonal
antibodies
demonstrate
the organization
of axons in the leech. J. Neurosci. 3: 369-375.
Hogg, N., M. Flaster, and B. Zipser (1983) Cross reactivities of
monoclonal
antibodies
between select leech neuronal and
epithelial tissues. J. Neurosci. Res. 9: 445-457.
Macagno, E. R. (1980) Number and distribution
of neurons in
leech segmental ganglia. J. Comp. Neurol. 190: 283-302.
Mann, K. H. (1962) Leeches (Hirudinea),
Pergamon
Press,
New York.
McKay, R., S. Hockfield, J. Johansen, L. Kleina, and I. Thompson (1982) Monoclonal
antibodies
as probes of the leech
nervous system. Sot. Neurosci. Abstr. 8: 714.
Stent, G. S., D. A. Weisblat, S. S. Blair, and S. L. Zackson
(1982) Cell lineage in the development
of the leech nervous
system. In Neuronal Development,
N.C. Spitzer, ed., pp. l44, Plenum Press, New York.
Stewart, R. R., E. R. Macagno,
and B. Zipser (1982) The
labeling patterns
of developing
leech CNS neurons with
monoclonal
antibodies raised against adult nervous tissue.
Sot. Neurosci. Abstr. 8: 15.
Weisblat, D. A. (1981) Development
of the nervous system. In
Neurobiology
of the Leech, K. J. Muller, J. G. Nicholls, and
G. S. Stent, eds., pp. 173-195, Cold Spring Harbor Laboratory, Cold Spring Harbor, NY.
Zipser, B. (1979a) Identifiable
neurons
controlling
penile
eversion in the leech. J. Neurophysiol.
42: 455-464.
Zipser, B. (197913) Voltage-modulated
membrane resistance in
coupled neurons. J. Neurophysiol.
42: 465-475.
Zipser, B. (1982) Complete distribution
patterns of neurons
with characteristic
antigens in the leech central nervous
system. J. Neurosci. 2: 1453-1464.
Zipser, B., and R. McKay (1981) Monoclonal
antibodies distinguish neurons in the leech. Nature 289: 549-554.
Zipser, B., S. Hockfield, and R. McKay (1981) Immunological
identification
of specific neurons. In Neurobiology
of the
Leech, K. J. Muller, J. G. Nicholls, and G. S. Stent, eds., pp.
235-247, Cold Spring Harbor Laboratory,
Cold Spring Harbor, NY.
Zipser, B., N. Hogg, J. Smart, and F. Hendrickson
(1982) Some
monoclonal
antibodies which recognize leech neurons also
distinguish
classes of peripheral
epithelium.
Sot. Neurosci.
Abstr. 8: 416.